You are not logged in.

#1 2026-06-07 23:03:37

mesaprotector
Member
Registered: 2024-03-03
Posts: 199

NVIDIA GPU periodically falls off bus

Hello all, me and my misbehaving NVIDIA hybrid laptop have returned after a year or so of it being cooperative.

Starting about a month ago, the dGPU falls off the bus (Xid 79) every once in a while. Usually this comes in the form of my going away for a bit and coming back to a black screen (normal) that won't resume on keypresses (not normal). But one time it happened during the process of shutting down so it isn't just when the system is idle.

If I terminate Xorg (and everything else) with SysRq+E then the display returns at the TTY. Changing TTYs while the screen is black doesn't help it come back though.

As far as I can tell from incomplete journals the first time this happened was May 3. The kernel upgraded to 7.0 on April 29 along with nvidia-open-dkms upgrading to 595.71.

Of note is that I've also had a couple weird freezes in Windows 11, starting around the same time, where the display freezes a couple seconds after I enter my password to log in. That makes me hope my GPU isn't just starting to die. Oh, and yes, I checked again, fast startup is disabled.

This issue looks similar and the time frame is similar as well but it could be something entirely different: https://bbs.archlinux.org/viewtopic.php?id=313284

Offline

#2 2026-06-08 06:19:37

seth
Member
From: Won't reply 2 private help req
Registered: 2012-09-03
Posts: 76,824

Re: NVIDIA GPU periodically falls off bus

That makes me hope my GPU isn't just starting to die.

first time this happened was May 3. The kernel upgraded to 7.0 on April 29 along with nvidia-open-dkms upgrading to 595.71.

https://wiki.archlinux.org/title/Arch_L … cific_date

https://bbs.archlinux.org/viewtopic.php?id=313284

Knowing this, I've currently alleviated the issue by forcing it into P0 via

sudo nvidia-smi -lgc 500,2100
sudo nvidia-smi -lmc 6001,6001

Have you tried similar?

Online

#3 2026-07-13 03:58:39

mesaprotector
Member
Registered: 2024-03-03
Posts: 199

Re: NVIDIA GPU periodically falls off bus

Apologies, I wasn't able to try most of what seth suggested, but here is some additional information before I have time to debug more thoroughly.

- ~90% of the times this happens are with the laptop unplugged
- has happened when waking up from screen being off, launching mpv, or on shutdown (doesn't happen consistently in any case/not reproducible)
- I did try keeping the dGPU in P0/P3/P5/P8, with no immediate problems (I didn't try doing this long-term).

Example of the journal at the start of a crash (the system is usually functional during the 65535ms countdown, as long as I don't ask for the dGPU to do anything):

Jul 12 23:35:01 Marojejy kernel: nvidia 0000:01:00.0: not ready 1023ms after resume; waiting
Jul 12 23:35:03 Marojejy kernel: nvidia 0000:01:00.0: not ready 2047ms after resume; waiting
Jul 12 23:35:05 Marojejy kernel: nvidia 0000:01:00.0: not ready 4095ms after resume; waiting
Jul 12 23:35:09 Marojejy kernel: nvidia 0000:01:00.0: not ready 8191ms after resume; waiting
Jul 12 23:35:18 Marojejy kernel: nvidia 0000:01:00.0: not ready 16383ms after resume; waiting
Jul 12 23:35:35 Marojejy kernel: nvidia 0000:01:00.0: not ready 32767ms after resume; waiting
Jul 12 23:36:09 Marojejy kernel: nvidia 0000:01:00.0: not ready 65535ms after resume; giving up
Jul 12 23:36:09 Marojejy kernel: NVRM: GPU at PCI:0000:01:00: GPU-951b6976-0ce6-589c-fc83-794df72b0d60
Jul 12 23:36:09 Marojejy kernel: NVRM: Xid (PCI:0000:01:00): 79, GPU has fallen off the bus.
Jul 12 23:36:09 Marojejy kernel: NVRM: GPU 0000:01:00.0: GPU has fallen off the bus.
Jul 12 23:36:09 Marojejy kernel: NVRM: krcRcAndNotifyAllChannels_IMPL: RC all channels for critical error 79.
Jul 12 23:36:09 Marojejy kernel: NVRM: _threadNodeCheckTimeout: API_GPU_ATTACHED_SANITY_CHECK failed!
Jul 12 23:36:09 Marojejy kernel: NVRM: _threadNodeCheckTimeout: API_GPU_ATTACHED_SANITY_CHECK failed!
Jul 12 23:36:09 Marojejy kernel: NVRM: _threadNodeCheckTimeout: API_GPU_ATTACHED_SANITY_CHECK failed!
Jul 12 23:36:09 Marojejy kernel: NVRM: _threadNodeCheckTimeout: API_GPU_ATTACHED_SANITY_CHECK failed!
Jul 12 23:36:09 Marojejy kernel: NVRM: _issueRpcAndWait: rpcSendMessage failed with status 0x0000000f for fn 78 sequence 6187!
Jul 12 23:36:09 Marojejy kernel: NVRM: nvCheckOkFailedNoLog: Check failed: GPU lost from the bus [NV_ERR_GPU_IS_LOST] (0x0000000F) returned from nvdEngineDumpCallbackHelper(pGpu, pPrbEnc, pNvDumpState, pEngineCallback) @ nv_debug_dump.c:273
Jul 12 23:36:09 Marojejy kernel: NVRM: _issueRpcAndWait: rpcSendMessage failed with status 0x0000000f for fn 78 sequence 6188!
Jul 12 23:36:09 Marojejy kernel: NVRM: nvCheckOkFailedNoLog: Check failed: GPU lost from the bus [NV_ERR_GPU_IS_LOST] (0x0000000F) returned from nvdEngineDumpCallbackHelper(pGpu, pPrbEnc, pNvDumpState, pEngineCallback) @ nv_debug_dump.c:273

A kernel warning that sometimes, but not always, accompanies the crash (is always printed twice when it does):

Jul 12 23:36:09 Marojejy kernel: ------------[ cut here ]------------
Jul 12 23:36:09 Marojejy kernel: rm_set_external_kernel_client_count(sp, NV_STATE_PTR(nvl), NV_FALSE) != NV_OK
Jul 12 23:36:09 Marojejy kernel: WARNING: nvidia/nv.c:5333 at nvidia_dev_put+0xbd/0xd0 [nvidia], CPU#5: InputThread/1688
Jul 12 23:36:09 Marojejy kernel: Modules linked in: nft_fib_ipv4 nft_ct nft_fib_ipv6 nft_fib wireguard libcurve25519 ip6_udp_tunnel udp_tunnel snd_seq_dummy rfcomm snd_hrtimer snd_seq snd_seq_device uinput snd_hda_codec_intelhdmi snd_hda_codec_hdmi xe snd_hda_codec_alc269 snd_hda_codec_realtek_lib drm_gpusvm_helper snd_hda_scodec_component drm_gpuvm snd_hda_codec_generic gpu_sched drm_exec drm_suballoc_helper ccm algif_aead des_generic libdes md4 algif_skcipher algif_hash af_alg bnep uvcvideo btusb btmtk videobuf2_vmalloc uvc btrtl videobuf2_memops videobuf2_v4l2 btbcm btintel videobuf2_common videodev bluetooth mc snd_sof_pci_intel_tgl snd_sof_pci_intel_cnl snd_sof_intel_hda_generic soundwire_intel snd_sof_intel_hda_sdw_bpt snd_sof_intel_hda_common snd_soc_hdac_hda snd_sof_intel_hda_mlink snd_sof_intel_hda soundwire_cadence snd_sof_pci snd_sof_xtensa_dsp snd_sof snd_sof_utils snd_soc_acpi_intel_match snd_soc_acpi_intel_sdca_quirks soundwire_generic_allocation snd_soc_sdw_utils snd_soc_acpi soundwire_bus snd_soc_sdca intel_rapl_msr
Jul 12 23:36:09 Marojejy kernel:  intel_uncore_frequency crc8 intel_uncore_frequency_common intel_tcc_cooling snd_soc_avs ip6t_REJECT x86_pkg_temp_thermal rtw89_8852be nf_reject_ipv6 rtw89_8852b intel_powerclamp snd_soc_hda_codec snd_hda_intel coretemp snd_hda_ext_core rtw89_8852b_common snd_soc_core xt_hl snd_hda_codec kvm_intel rtw89_pci snd_hda_core joydev mousedev snd_intel_dspcfg snd_compress ip6t_rt ac97_bus rtw89_core kvm snd_intel_sdw_acpi iTCO_wdt r8169 snd_pcm_dmaengine snd_hwdep irqbypass spd5118 hid_multitouch mei_hdcp mei_pxp intel_pmc_bxt realtek snd_pcm aesni_intel snd_ctl_led ipt_REJECT mac80211 ucsi_acpi nf_reject_ipv4 lenovo_wmi_hotkey_utilities phy_package gf128mul snd_timer lenovo_wmi_other lenovo_wmi_gamezone xt_comment spi_nor typec_ucsi rapl lenovo_wmi_helpers vfat snd mdio_devres roles nvidia_drm(OE) intel_cstate legion_laptop(OE) fat firmware_attributes_class lenovo_wmi_events i915 lenovo_wmi_capdata cfg80211 wmi_bmof nvidia_wmi_ec_backlight nft_limit nvidia_modeset(OE) processor_thermal_device_pci i2c_hid_acpi
Jul 12 23:36:09 Marojejy kernel:  intel_uncore soundcore libphy mtd xt_limit i2c_i801 ideapad_laptop libarc4 pcspkr processor_thermal_device typec mei_me drm_ttm_helper i2c_hid drm_buddy sparse_keymap processor_thermal_wt_hint i2c_smbus ttm mdio_bus i2c_mux mei thunderbolt intel_pmc_core xt_addrtype rfkill i2c_algo_bit platform_temperature_control elants_i2c xt_tcpudp drm_display_helper processor_thermal_soc_slider pmt_telemetry processor_thermal_rfim cec processor_thermal_rapl pmt_discovery xt_conntrack intel_rapl_common intel_gtt pmt_class video processor_thermal_wt_req nf_conntrack processor_thermal_power_floor int3400_thermal intel_pmc_ssram_telemetry processor_thermal_mbox nf_defrag_ipv6 int3403_thermal acpi_thermal_rel wmi pinctrl_tigerlake int340x_thermal_zone acpi_pad intel_vsec igen6_edac nf_defrag_ipv4 nft_compat mac_hid serio_raw x_tables nf_tables nfnetlink nvidia_uvm(OE) nvidia(OE) i2c_dev pkcs8_key_parser crypto_user aead nvme spi_intel_pci spi_intel nvme_core nvme_keyring intel_lpss_pci intel_lpss nvme_auth idma64
Jul 12 23:36:09 Marojejy kernel: CPU: 5 UID: 1000 PID: 1688 Comm: InputThread Tainted: G        W  OE       7.1.3-arch1-2 #1 PREEMPT(full)  c2fc2df4661dd5176652c8b9c2179d3924a4ac3a
Jul 12 23:36:09 Marojejy kernel: Tainted: [W]=WARN, [O]=OOT_MODULE, [E]=UNSIGNED_MODULE
Jul 12 23:36:09 Marojejy kernel: Hardware name: LENOVO 82XV/LNVNB161216, BIOS LZCN32WW 09/23/2023
Jul 12 23:36:09 Marojejy kernel: RIP: 0010:nvidia_dev_put+0xbd/0xd0 [nvidia]
Jul 12 23:36:09 Marojejy kernel: Code: 41 5d e9 46 2e 96 f3 5b 48 c7 c7 10 37 80 c0 5d 41 5c 41 5d e9 34 2e 96 f3 31 d2 48 89 de 4c 89 e7 e8 97 a2 16 00 85 c0 74 cf <0f> 0b eb cb 0f 1f 40 00 66 66 2e 0f 1f 84 00 00 00 00 00 90 90 90
Jul 12 23:36:09 Marojejy kernel: RSP: 0018:ffffd03a05f37998 EFLAGS: 00010202
Jul 12 23:36:09 Marojejy kernel: RAX: 0000000000000026 RBX: ffff8c2a17508000 RCX: 0000000000000000
Jul 12 23:36:09 Marojejy kernel: RDX: ffffd03a05f37928 RSI: 0000000000000296 RDI: ffffd03a05f378e8
Jul 12 23:36:09 Marojejy kernel: RBP: ffff8c2a17508730 R08: ffff8c2b25919540 R09: 0000000000000000
Jul 12 23:36:09 Marojejy kernel: R10: fffff1d308964640 R11: ffff8c2a00042200 R12: 0000000000000000
Jul 12 23:36:09 Marojejy kernel: R13: 0000000000000000 R14: 0000000000000000 R15: ffff8c2a22880000
Jul 12 23:36:09 Marojejy kernel: FS:  0000000000000000(0000) GS:ffff8c2de9691000(0000) knlGS:0000000000000000
Jul 12 23:36:09 Marojejy kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jul 12 23:36:09 Marojejy kernel: CR2: 00007f65c4cfece8 CR3: 0000000146658006 CR4: 0000000000f72ef0
Jul 12 23:36:09 Marojejy kernel: PKRU: 55555554
Jul 12 23:36:09 Marojejy kernel: Call Trace:
Jul 12 23:36:09 Marojejy kernel:  <TASK>
Jul 12 23:36:09 Marojejy kernel:  nvidia_close+0x26f/0x300 [nvidia ee1f2b6525a49b481124018dc4f27eb84e6d4a15]
Jul 12 23:36:09 Marojejy kernel:  __fput+0xf6/0x2d0
Jul 12 23:36:09 Marojejy kernel:  task_work_run+0x66/0xa0
Jul 12 23:36:09 Marojejy kernel:  do_exit+0x304/0xba0
Jul 12 23:36:09 Marojejy kernel:  ? __pfx_futex_wake_mark+0x10/0x10
Jul 12 23:36:09 Marojejy kernel:  do_group_exit+0x2d/0xc0
Jul 12 23:36:09 Marojejy kernel:  get_signal+0x85c/0x910
Jul 12 23:36:09 Marojejy kernel:  arch_do_signal_or_restart+0x77/0x2d0
Jul 12 23:36:09 Marojejy kernel:  exit_to_user_mode_loop+0x95/0x6a0
Jul 12 23:36:09 Marojejy kernel:  do_syscall_64+0x21d/0x660
Jul 12 23:36:09 Marojejy kernel:  ? __pfx_ep_autoremove_wake_function+0x10/0x10
Jul 12 23:36:09 Marojejy kernel:  ? __x64_sys_epoll_wait+0x70/0x120
Jul 12 23:36:09 Marojejy kernel:  ? switch_fpu_return+0x5e/0x100
Jul 12 23:36:09 Marojejy kernel:  ? do_syscall_64+0x252/0x660
Jul 12 23:36:09 Marojejy kernel:  ? __wake_up_sync_key+0x43/0x60
Jul 12 23:36:09 Marojejy kernel:  ? anon_pipe_write+0x2bc/0x680
Jul 12 23:36:09 Marojejy kernel:  ? apparmor_file_permission+0x44/0xb0
Jul 12 23:36:09 Marojejy kernel:  ? security_file_permission+0x5d/0x130
Jul 12 23:36:09 Marojejy kernel:  ? vfs_write+0x454/0x570
Jul 12 23:36:09 Marojejy kernel:  ? ksys_write+0xdd/0x110
Jul 12 23:36:09 Marojejy kernel:  ? do_syscall_64+0x623/0x660
Jul 12 23:36:09 Marojejy kernel:  ? ksys_read+0xdd/0x110
Jul 12 23:36:09 Marojejy kernel:  ? do_syscall_64+0x623/0x660
Jul 12 23:36:09 Marojejy kernel:  ? do_syscall_64+0x5f/0x660
Jul 12 23:36:09 Marojejy kernel:  entry_SYSCALL_64_after_hwframe+0x76/0x7e
Jul 12 23:36:09 Marojejy kernel: RIP: 0033:0x7f6848c945be
Jul 12 23:36:09 Marojejy kernel: Code: Unable to access opcode bytes at 0x7f6848c94594.
Jul 12 23:36:09 Marojejy kernel: RSP: 002b:00007f682cdfde08 EFLAGS: 00000282 ORIG_RAX: 00000000000000ca
Jul 12 23:36:09 Marojejy kernel: RAX: fffffffffffffe00 RBX: 0000559b3cd10420 RCX: 00007f6848c945be
Jul 12 23:36:09 Marojejy kernel: RDX: 0000000000000002 RSI: 0000000000000080 RDI: 0000559b21723e60
Jul 12 23:36:09 Marojejy kernel: RBP: 00007f682cdfde20 R08: 0000000000000000 R09: 0000000000000000
Jul 12 23:36:09 Marojejy kernel: R10: 0000000000000000 R11: 0000000000000282 R12: 0000000000000001
Jul 12 23:36:09 Marojejy kernel: R13: 0000559b3c99b520 R14: 0000000000000026 R15: 0000000000000001
Jul 12 23:36:09 Marojejy kernel:  </TASK>
Jul 12 23:36:09 Marojejy kernel: ---[ end trace 0000000000000000 ]---
Jul 12 23:36:09 Marojejy systemd[1537]: kitty-1705-0.scope: Consumed 8.272s CPU time over 1h 44min 38.366s wall clock time, 90.4M memory peak.
Jul 12 23:36:09 Marojejy kernel: ------------[ cut here ]------------

Last edited by mesaprotector (2026-07-13 04:00:37)

Offline

Board footer

Powered by FluxBB